1Dalhousie University, Department of Oceanography, Halifax, Canada
2Department de Biologié, Université Laval, Québec, Québec, Canada
3Demersal and Benthic Sciences Division, Maurice-Lamontagne Institute, Fisheries and Oceans Canada, Mont-Joli, Québec, Canada
4Arctic and Aquatic Research Division, Fisheries and Oceans Canada, Winnipeg, Manitoba, Canada
5Institute of Marine Research, His, Norway
6Center of Earth Observation Science, University of Manitoba, Winnipeg, Manitoba, Canada
7Centre for Arctic Knowledge and Exploration, Canadian Museum of Nature, Ottawa, ON, Canada

1 Introduction

1.1 Current knowledge

1.1.1 Present

  • Evidence suggests that many Arctic coasts should support seaweed
  • In Canada, kelp has been reported and documented along Arctic and subarctic coastlines
  • However, baseline measures of the extent of kelp communities are missing in much of the region

1.1.2 Future

  • Rapid environmental changes, such as declining sea ice, increased ocean temperatures, and freshwater inputs are occurring along Canadian coasts
  • Research suggests northern expansion of kelp forests with climate change
  • Therefore, the relationships between environmental factors and the presence of kelp forests in the Canadian Arctic are critical to understand

1.1.3 Existing database

  • Canadian Museum of Nature (CANA)
  • Global Biodiversity Information Facility (GBIF; www.gbif.org)
  • Ocean Biogeographic Information System (OBIS; www.obis.org)
  • Literature (Borum et al., 2002; Filbee-Dexter et al., 2019; Hop et al., 2016; Ronowicz et al., 2020; Schoenrock et al., 2018)
  • ArcticKelp project

1.1.4 ArcticKelp project

  • Dive research conducted throughout the Canadian Arctic in 2014 - 2019
    • 5 - 20 m photograph quadrats

1.1.4.1 Campaigns

1.1.4.2 Mean cover

1.2 Questions

  • Is it possible to model the distribution (suitability + abundance) of different functional groups of kelps in the Arctic given our current knowledge?
    • Laminariales (Laminaria sp. + Sacharina sp.)
    • Agarum
    • Alaria
  • How accurate are the models?
  • Which environmental variables are the most important?
  • What do present + future distributions look like?

2 Methods

2.1 Study region

2.2 Environmental data


(Assis et al., 2018; Tyberghein et al., 2012)

2.2.1 Bio-ORACLE

  • Geophysical, biotic, and abiotic environmental variables
  • Collection from many different datasets
  • Surface and benthic coverage
  • Data from 2000 - 2014 for most
  • Single values per pixel: min, mean, max, and range for most
  • 5 arcdegree spatial resolution (~9.2 km at the equator)

2.2.2 Variables (32)

  • Temperature
  • Salinity
  • Ice thickness (surface only)
  • Current velocity
  • Photosynthetically active radiation (PAR; surface only)
  • Dissolved oxygen
  • Iron
  • Nitrate
  • Phosphate

2.2.3 Final variables (8)

  • Bottom temperature; long-term minimum
  • Bottom temperature; long-term maximum
  • Surface temperature; long-term maximum
  • Surface salinity; long-term maximum
  • Ice thickness; long-term minimum
  • Bottom iron; long-term maximum
  • Bottom phosphate; long-term maximum
  • Bottom current velocity; long-term minimum

2.2.4 Future variables (6)

  • Bottom temperature; long-term minimum
  • Bottom temperature; long-term maximum
  • Surface temperature; long-term maximum
  • Bottom salinity; long-term maximum
  • Ice thickness; long-term minimum
  • Bottom current velocity; long-term minimum

2.3 Ensemble model (suitability)

  • Presnce data
  • Ensemble performed with default BIOMOD2 R package settings (Thuiller et al., 2020)
  • Models: MAXENT (Phillips), GLM, ANN, RF, GAM (Goldsmit et al., 2020)
  • Random pseudo-absence (PA); 1000 points; 5 repetitions
  • 70/30 train test split
  • 0.7 TSS cutoff
  • Modeled for entire Arctic ecoregion
  • Results cropped to Eastern Canadian Arctic

2.4 Random forest model (abundance)

  • Percent cover data
  • 200 trees; 1000 repetitions
  • 70/30 train test split
  • Modeled only for Eastern Canadian Arctic

3 Results

3.1 Ensemble

Confidence

Laminariales

digitata

solidungula

latissima

Agarum

Alaria

Variables

Laminariales

digitata
Data layer Importance
Temperature (bottom, max.) 0.59
Temperature (surface, max.) 0.32
Iron (bottom, max.) 0.24
Temperature (bottom, min.) 0.23
Phosphate (bottom, max.) 0.23
Ice thickness (surface, min.) 0.19
Salinity (surface, max.) 0.17
Current velocity (bottom, min.) 0.08
solidungula
Data layer Importance
Temperature (surface, max.) 0.58
Temperature (bottom, min.) 0.29
Iron (bottom, max.) 0.24
Phosphate (bottom, max.) 0.23
Temperature (bottom, max.) 0.20
Ice thickness (surface, min.) 0.15
Salinity (surface, max.) 0.13
Current velocity (bottom, min.) 0.10
latissima
Data layer Importance
Temperature (bottom, max.) 0.37
Temperature (surface, max.) 0.34
Iron (bottom, max.) 0.16
Temperature (bottom, min.) 0.15
Ice thickness (surface, min.) 0.08
Phosphate (bottom, max.) 0.08
Salinity (surface, max.) 0.06
Current velocity (bottom, min.) 0.02

Agarum

Data layer Importance
Temperature (bottom, max.) 0.37
Ice thickness (surface, min.) 0.30
Temperature (surface, max.) 0.28
Salinity (surface, max.) 0.26
Temperature (bottom, min.) 0.14
Phosphate (bottom, max.) 0.14
Iron (bottom, max.) 0.11
Current velocity (bottom, min.) 0.05

Alaria

Data layer Importance
Temperature (bottom, max.) 0.43
Temperature (surface, max.) 0.20
Ice thickness (surface, min.) 0.17
Phosphate (bottom, max.) 0.14
Salinity (surface, max.) 0.12
Temperature (bottom, min.) 0.11
Iron (bottom, max.) 0.07
Current velocity (bottom, min.) 0.02

Projections

Laminariales

digitata

solidungula

latissima

Agarum

Alaria

3.2 Random Forest

Confidence

Laminariales

Agarum

Alaria

Variables

Laminariales

Data layer % Increase MSE
Temperature (bottom, min.) 55
Salinity (surface, max.) 52
Temperature (surface, max.) 52
Iron (bottom, max.) 47
Phosphate (bottom, max.) 46
Current velocity (bottom, min.) 42
Temperature (bottom, max.) 42
Ice thickness (surface, min.) 0

Agarum

Data layer % Increase MSE
Temperature (bottom, min.) 157
Iron (bottom, max.) 132
Salinity (surface, max.) 112
Temperature (surface, max.) 95
Phosphate (bottom, max.) 94
Current velocity (bottom, min.) 68
Temperature (bottom, max.) 66
Ice thickness (surface, min.) 0

Alaria

Data layer % Increase MSE
Salinity (surface, max.) 41
Temperature (bottom, min.) 35
Phosphate (bottom, max.) 26
Current velocity (bottom, min.) 26
Temperature (surface, max.) 24
Temperature (bottom, max.) 21
Iron (bottom, max.) 17
Ice thickness (surface, min.) 0

Projections

Laminariales

Agarum

Alaria

4 Conclusions

  • Confidence in MAXENT models in the ensembles is low
  • Random forests tend to underestimate large values, and overestimate small ones
  • Temperature is usually one of the most important variables for both models
  • Ice is always the least important for the random forest
  • Laminariales are projected to decline in the future in both models
  • Agarum and Alaria show a mix of increase and decline in both models
  • These projections provide a good platform for deciding future sampling locations

5 Further work

  • Improve MAXENT models
  • More experimentation with variable choice
  • Comparisons of ensemble and random forest outputs

6 Acknowledgements

  • This research was undertaken thanks in part to funding from the Canada First Research Excellence Fund, through the Ocean Frontier Institute.

7 Questions ?

Alaria

Linear regression

Laminariales

Agarum

Alaria

References

Assis, J., Tyberghein, L., Bosch, S., Verbruggen, H., Serrão, E. A., and De Clerck, O. (2018). Bio-oracle v2. 0: Extending marine data layers for bioclimatic modelling. Global Ecology and Biogeography 27, 277–284.

Borum, J., Pedersen, M., Krause-Jensen, D., Christensen, P., and Nielsen, K. (2002). Biomass, photosynthesis and growth of laminariasaccharina in a high-arctic fjord, ne greenland. Marine Biology 141, 11–19.

Filbee-Dexter, K., Wernberg, T., Fredriksen, S., Norderhaug, K. M., and Pedersen, M. F. (2019). Arctic kelp forests: Diversity, resilience and future. Global and Planetary Change 172, 1–14.

Goldsmit, J., McKindsey, C. W., Schlegel, R. W., Stewart, D. B., Archambault, P., and Howland, K. L. (2020). What and where? Predicting invasion hotspots in the arctic marine realm. Global change biology 26, 4752–4771.

Hop, H., Kovaltchouk, N. A., and Wiencke, C. (2016). Distribution of macroalgae in kongsfjorden, svalbard. Polar Biology 39, 2037–2051.

Ronowicz, M., Włodarska-Kowalczuk, M., and Kukliński, P. (2020). Glacial and depth influence on sublittoral macroalgal standing stock in a high-arctic fjord. Continental Shelf Research 194, 104045.

Schoenrock, K., Stachnik, Ł., Vad, J., Kamenos, N., Pearce, D., Rea, B., et al. (2018). DISTRIBUTION of benthic communitites in a fjord-marine system in southwestern greenland, with a focus on algal dominated habitats. in АРКТИЧЕСКИЕ исследования: ОТ экстенсивного освоения к комплексному развитию, 220–223.

Thuiller, W., Georges, D., Engler, R., and Breiner, F. (2020). Biomod2: Ensemble platform for species distribution modeling. Available at: https://CRAN.R-project.org/package=biomod2.

Tyberghein, L., Verbruggen, H., Pauly, K., Troupin, C., Mineur, F., and De Clerck, O. (2012). Bio-oracle: A global environmental dataset for marine species distribution modelling. Global ecology and biogeography 21, 272–281.